Tadayuki Kodama (NAOJ)

Yusei Koyama (NAOJ), Ken-ichi Tadaki (NAOJ), Masao Hayashi (Univ of Tokyo), Ichi Tanaka (Subaru),

Rhythm Shimakawa (Subaru)　　　　　　　　　　　　　　　　　　　　　　　　　　　　　

(銀河団の物理@東京理科大学, 2013/12/28)

A galaxy cluster RXJ0152 at z=0.83 (Subaru/Suprime-Cam)

光赤外で探る銀河団銀河の形成と進化 Optical-IR Views of Formation and Evolution of Galaxy Clusters

z = 3

Nature? (intrinsic) earlier galaxy formation and

evolution in high density regions

Nurture? (external) galaxy-galaxy interaction/mergers,

gas-stripping

Spirals Lenticulars

Ellipticals

Star-forming (young) No/little SF

(old)

Morphology- (SFR-) density relation (Dressler 1980)

z~0

log surface density (Mpc-2)

Origin of the cosmic habitat segregation

Star formation

The peak epoch of galaxy formation: 1<z<3 (6>Tcos(Gyr)>2)

Fan et al. (2006)

AGN (QSO)

Hopkins and Beacom (2006)

“COSMIC NOON”

4 narrow-band filters 7 narrow-band filters

FWHMs correspond to ±1500-2000km/s

Subaru Wide-Field Survey of Narrow-Band Emitters ([OII], [OIII] and Hα) at 0.4<z<2.5

MOIRCS FoV

7’x4’ Cluster N-body+SAM simulation

(Yahagi et al. 2005; νGC)

z = 2

Suprime-Cam (optical; 34’x27’) MOIRCS (NIR; 7’x4’)

20×20 Mpc2 box (co-moving)

A typical spectrum of star forming galaxy

Kinney et al. (1996)

[OII] [OIII]

[Hα]

[SII] [Hβ]

[SIII]

It consists of blue stellar continuum and many emission lines from ionized gas.

18 nights for imaging, >15 nights for spectroscopy

“MAHALO-Subaru” MApping HAlpha and Lines of Oxygen with Subaru

Unique sample of NB selected SF galaxies across environments and cosmic times

z~2 cluster

z~2 field

z~1.5 cluster

z<1 cluster

(SFRHα > 0.3M☉/yr) Hα (6563A)@z=0.4

NB921 z’

Flux ratio between narrow-band (NB921) and

broad-band (z’)

Koyama et al. (2011)

Magnitudes in NB

How narrow-band imaging survey works to sample star forming galaxies at high-z

radio  galaxy NB2315

dense  clump

Hayashi et al. (2012)

Ks

Discovery of a Prominent Star-Bursting Proto-Cluster at z~2.5

Hα imaging with MOIRCS/NB2315 3.4 hrs, 0.3-0.4” seeing

~20x denser than the general field. Mean separation between galaxies is ~150kpc.

1.5Mpc away from the RG

USS1558-003 (z=2.53)

68 Hα emitters (HAEs) are detected.

Hayashi et al. (2012)

Red Hα emitters (dusty starbursts) tend to favor higher density regions!

Koyama et al. (2012)

Spatial distributions of HAEs in proto-clusters at z~2

★

★

★　RG

Blue HAE ■ Red HAE ■

clump-1

clump-2

clump-3

~10Mpc on a side (co-m

oving)

cluster core Figure 7: The kinematical structures of Hα emitters in (a) left: PKS1138 and (b) right: USS1558. Grey dotsrepresent Hα emitter candidates detected in our NB imaging (K13 and H12). Diamonds show the membersnewly confirmed in this study. Triangles are the confirmed Hα and Lyα emitters (LAE) in the previousworks (Kurk et al., 2000, 2004; Croft et al., 2005; Doherty et al., 2010). Blue and red symbols are separatedby blue- and red-shifted galaxies, respectively, relative to the RGs (star). We make groups such as PKS1138C1 and USS1558 C1–C3, and they are shown by grey dashed circles. Solid and dashed black circles indicateR200 and 0.5×R200, respectively.

Table 7: (1) Cluster name as shown in Fig. 4, (2) confirmed number in this work, (3) mean peculiar velocity,(4) velocity dispersion, and (5) R200 and (6) cluster mass (Mcl) calculated from Finn et al. (2005).

Cluster Confirm 〈∆V〉 σcl R200 MclName [km/s] [km/s] [Mpc] [1014M$]

(1) (2) (3) (4) (5) (6)PKS1138 all 27 −41 8824 — —PKS1138 C1 9 +9 6832 0.3712 1.20USS1558 all 37 −164 841 — —USS1558 C1 7 +426 508 0.236 0.42USS1558 C2 13 −1134 660 0.306 0.92USS1558 C3 6 −956 319 0.148 0.10

4 estimated including the literature

27

Environmental dependence in the formation phase of early-type galaxies?

•  Mode of star formation? (starburst, dusty) Location on the SFR-M* diagram SFE=SFR / f(gas) SFR(IR) / SFR(UV) or SFR(Hα) / SFR(UV) •  Internal structure? (clumpy, central burst) HST images of USS1558 proto-cluster (z=2.53) Hα imaging w/ AO, IFU spectroscopy, ALMA

Daddi et al. (2007)

“Main Sequence” of Star Forming Galaxies at z~2

SFR vs. M* (~proportional). Large scatter. SMG is up-scattered by ~×10. There are two modes of star formation: Normal mode and Burst mode

Good relation, but a considerable scatter!

!"#$%&$'()(%)"*+),-./-0

!%"1)2&13"%#&$)(24#5#($56)7"1#"%#&$8)#$)%9()!:;<=>)?*"$(

!"#$)@!)5&$%($%8)"$A)8%"1)2&13"%#&$)(24#5#($56)7"16)!"#$%%)%9()=!)

COLDGASS project

SDSS galaxies IRAM 30-m

SFE and fgas both go up with Δ(MS) !

Saintonge et al. (2012)

Star formation efficiency (SFE) and molecular gas fraction (fgas) as a function of deviation from the main sequence (ΔMS)

SFE

fgas

SFE = SFR / Mgas

fgas=Mgas/(Mgas+Mstar)

(sSFR = SFR / Mstar)

Wuyts et al. (2011)

Starburst galaxies (high-sSFR; up-scattered from the MS)

Nucleated starburst regions are compact and hence have large dust extinction, probably as a result of gas inflow due to galaxy-galaxy interactions/mergers.

are dustier (higher SFR(IR) / SFR(UV) ratios)

Passive

Star-forming

Starburst (mergers?)

fast track

late track

Stellar Mass (M*)

Sta

r For

mat

ion

Rat

e (S

FR)

Schematic diagram of evolution on SFR-M* (Main Sequence)

(Koyama et al. 2013a)

SF galaxies in the proto-cluster at z~2 follow the same “main sequence” as

the field one.

However,

the galaxy distribution on the sequence depends on

environment in a sense that the proto-cluster tends to contain more massive/high-SFR galaxies than the field.

Hα-selected sample only

M*-dependent dust correction for Hα is applied. (Garn & Best 2010)

Environmental Dependence of the Main-Sequence? à accelerated galaxy formation in proto-cluster (PKS1138) at z~2 �

Environmental dependence in the formation phase of early-type galaxies?

•  Mode of star formation? (starburst, dusty) Location on the SFR-M* diagram SFE=SFR / f(gas) SFR(IR) / SFR(UV) or SFR(Hα) / SFR(UV) •  Internal structure? (clumpy, central burst) HST images of USS1558 proto-cluster (z=2.53) Hα imaging w/ AO, IFU spectroscopy, ALMA

HST images (V606,I814,H160) from the CANDELS survey less massive clumpy galaxies

（Mstar<1010M◉）

0.9

0.8 0.6

0.5

0.8 0.7

0.9 0.6

0.5 0.6 0.4

0.4

2.2

1.2

1.7

1.4

1.1

0.2

1.7

0.9

1.6 2.0 1.9

1.0

1.8

1.3 1.0

0.7

2.0

0.9

massive clumpy galaxies （Mstar=1010-11M◉）

Tadaki et al. (2013b)

~40% of HAEs at z~2 show clumpy (or merger) structures

Clumpy Structure is Common

colours (I814-H160) of individual clumps are shown with red numbers

Massive galaxies tend to have a red clump near the mass center, which may be hosting a central dusty starburst and forming a bulge eventually ! Environmental dependence of the clumpiness and the clump colours is expected, and should be tested with up-coming HST imaging of the USS1558 proto-cluster.

3”

3” (~25kpc)

Dekel et al. (2009, Nature)

Goerdt et al. (2010)

320 kpc

10 kpc

Efficient gas supply to form a massive galaxy on a short time scale at high-z.

Rapid gas accretion forms a gas rich disk which

becomes gravitationally unstable and fragmented.

“Cold Streams” along filaments (Inflow)

cooling radiation of Lyα

6

Fig. 2.— Same as Figure 1 for galaxy G2 (medium mass). Detailed sequences and movies of our fiducial models are available in Perretet al. (2013a).

Bournaud et al. (2013)

Medium-mass galaxy

2C

Stellar feedback (photo-ionization, radiation pressure, and supernova) are fully considered.

Numerical simulation of clumpy galaxies

180Myr 300Myr

420Myr

640Myr 760Myr

540Myr

High-z galaxies are gas rich due to massive gas accretion through cold streams. Gas rich disks are fragmented to clumps due to gravitational instability. Clumps migrate towards a galactic center due to dynamical friction and probably make a bulge of a disk galaxy.

Mdyn = 3.5 × 1010 M◉

Environmental dependence?

Galaxy Anatomy at z~2 as a function of Environment

AO+NB imaging and IFU SF regions, AGN, kinematics ALMA gas distribution, SF mode, kinematics

HST imaging size, morphology, clumpiness

Galaxy mergers? Tidal signatures? Nucleated dusty SB? AGN? Disordered kinematics? Central outflow? Clump migration? SF clumps? Ordinary rotating disk? 　　　　　　　　　　　　　　 Outflows from clumps? Secular evolution? Exponential disk-wide star formation? Well-ordered rotational disk?

All these new exciting data are up-coming within a year or so! SINS

“GRACIAS-ALMA” GRAphing CO Intensity And Submm with ALMA

Mapping/resolving gas and dust contents at the peak epoch of galaxy formation

CO line @ Band-3 (~100GHz) SFR~50M☉/yr (2.7hrs, 5σ) Dust continuum@ Band-6-9 (450 µm–1.1 mm)

< 0.1” (<1kpc) : centralized, disturbed or disk-wide gas distribution? < 50km/s: gas in-/out-flow, rotating disk or disturbed motions?

Internal structures:

Cycle-2 sensitivities

Spatial resolution: 0.01-0.1” (ßà 0.1-1kpc) (0.18-0.4” in cycle-2)

SFR~15M☉/yr (50min, 5σ) @1<z<3

Band-6 (dust)

Band-3 (CO)

Clusters are efficient targets for ALMA especially at Band-3 as multiple targets can be observed by a single pointing (1’).

USS1558 proto-cluster (z=2.53)

Chandra 100ks X-ray data (Martini et al.) will also be taken soon.

(AGN fraction, distribution)

HST images (Hayashi et al.) will be taken soon.

(Clumpy fraction, size evolution)

HSC-Deep (27 deg2) Hybrid Cluster Survey to z~1.7

HSC NB

filters

“red sequence” survey (passive galaxies) “blue sequence” survey (NB emitters) (or “phot-z”)

~10 NB-selected clusters at z~1.5 ~200 clusters (>1014M◉) at z~1 （per Δz=1)

Large-scale (~20Mpc) structures at z~1.5 unveiled by [OII]λ3727 emitter survey with S-Cam

Tadaki et al. (2012)

Hayashi et al. (2011)

●

(Hilton+09)

(Papovich+10)

XCS2215(z=1.46)

ClGJ0218(z=1.62)

NB973

NB912

Expected number of galaxy clusters

survey area cluster number / (Δz=1)

HSC-Deep 27 deg2 200 (>1014 M◉)　at z=1

　(AB=27) 6 (>1014.5M◉)　at z=1

HSC-Wide 1400 deg2 10,000 (>1014 M◉)　at z=1

(AB=26) 300 (>1014.5M◉)　at z=1

Euclid-Deep 20 deg2 6 (>1014 M◉)　at z=2

(AB=26) 0.015 (>1014.5M◉)　at z=2

Euclid-Wide 6000 deg2 1,800 (>1014 M◉)　at z=2

(AB=24) 4 (>1014.5M◉)　at z=2

Ref: HSC-SSP proposal

NB Filter

Wavelength (µm)

Redshift (Hα, OIII)

SFR (Msun/yr)

Number / FoV

NB217 (same as SWIMS-18)

2.17 2.3, 3.3 7 ~1500

NB284 2.84 3.3, 4.7 20 ~200

NB335 3.35 4.1, 5.7 30 ~150

NB441 4.41 5.7, 7.8 90 ~20

NB497 4.97 6.6, 8.9 250 ??

6hrs exp. (5σ), 1mag extinction of Hα

NB filters at 2.5 ~ 5 µm will be the key for WISH !

WISH Space Telescope WISH is a 1.5m diameter space telescope for 1-5µm (NIR) with a FoV of 0.24 sq. deg.

Proposing NB filters for WISH

WISH-7 Survey

•  「7」 sq. deg. ~7×107 Mpc3 / (Δz=1) ~10 progenitors of Coma class clusters at every z

NB: 3.5-7×106 Mpc3 / (Δz=0.05-0.1) ~ 1 progenitor of Coma class cluster per filter

30 WISH pointings (0.24 sq. deg. / FoV) •  Exposure times needed 　　　BB: 1 hrs × 6 filters x 30 p = 180 hrs 　　　NB: 6 hrs × 4 filters x 30 p = 720 hrs 　　900 hrs in total (or 1200 hrs including overheads)

for 1.5<z<8

Summary •  Mahalo-Subaru is mapping out star formation activities across

cosmic times and environment, covering the peak epoch of galaxy formation and evolution (1<z<3).

•  Enhanced star forming activities in cluster cores at z~2

•  More massive SFGs in proto-clusters, indicating the biased, accelerated galaxy formation in clusters.

•  The mode of star formation (e.g. dusty starburst) may depend on environment.

•  Clumpy nature of SFGs at z~2 (especially the red clumps) maybe closely related to a bulge formation. We expect some environmental dependence in internal structures of SFGs.

•  HSC will make a large, complete sample of 10K clusters to z~1.7. WISH will extend it to z~6-8.